Well, we’ve just finished Release 2 of Calais – so it’s time to start talking about what’s in the pipeline for our next few months. This isn’t a roadmap – but it should give you a general idea of the areas we’ll be focusing on. As always, we welcome your input and ideas.
We hate vaporware – so there are a few things we’re working on that we’re not going to talk about until they’re ready. A few things we know will be happening over the next 90-120 days…
Expanding breadth and depth of our metadata generation capabilities
R2 was a milestone for us and the technical platform underlying Calais. In this release we began to take advantage of some of the many open data assets to deploy new entities such as sporting events, entertainment awards and others. This was our trial run – which resulted in nearly a dozen new entity types. We don’t believe in a pure lexicon-driven approach (it delivers unreliable results in too many cases), but the combination of expanded lexicons with a great NLP (Natural Language Processing) wrapper is delivering high-quality results very fast. We tried it. It worked. It’s going to accelerate – and we’re very open to hearing about the types of entities you’d like to see added in the future.
Linked Data
We could write on this topic for a few hundred pages – but we won’t. One significant area of development work for us is going to be creating the connections that allow you to move from what Calais extracts to the wider world of linked data assets. We’re not going to share the details right now, but we think it’s going to open up a whole new world of potential applications.
More Integration Tools
We’re going to keep working on building tools to make Calais more accessible and easier to integrate. Two of those tools for the mid-term time frame are the hooks necessary to integrate Calais with Yahoo Pipes and Microsoft Popfly.
More plugins, extensions, modules to augment popular tools
In R2 we released our WordPress plugin and others developed modules for Drupal and a UIMA annotator. We’ll continue this effort by releasing additional plugins and updated and improved versions of the ones we have.
As always – thanks for the support, suggestions, blog mentions and the occasional zing. It’s a wild ride.
Tom
Trackback URL for this post:
- Tom's blog
- Login or register to post comments





